The efficiency of endoscopic pancreatic function testing is optimized using duodenal aspirates at 30 and 45 minutes after intravenous secretin

OBJECTIVES: The endoscopic secretin test (ePFT) takes an hour to perform, limiting its practicality for routine use. We sought to determine which timed endoscopic collections best predict the peak bicarbonate concentration from full hour-long test. METHODS: A cross-sectional study was performed of all ePFTs performed at our institution from 2000 to 2004. We compared 7 theoretical shortened tests [4-single collections (15, 30, 45, or 60 minutes after secretin) and 3-dual collections (15/30, 30/45, 45/60 minutes after secretin)] to the full test and to each other using multivariate linear and Deming regression models. RESULTS: The bicarbonate results of 240 ePFTs were analyzed. The dual 30/45-minute collection was the optimal method based on the closest agreement with the full test (94%). Deming regression reveals the 30/45 estimate to be the best combination from an analytic perspective, owing to its lack of significant proportional bias and minimal constant bias (5.1 mEq/L) as compared to the reference test. Use of the 30/45 method with the adjusted cutpoint 75 mEq/L produces very good specificity (93%) for ruling in exocrine insufficiency. CONCLUSIONS: A dual timed aspiration at 30 and 45 minutes can be used to screen for pancreatic exocrine insufficiency in patients with abdominal pain, simplifying the use of the ePFT in clinical practice.